Association Between Cardiac Rehabilitation and 1-Year Mortality by Frailty Level in Medicare Beneficiaries

Frailty before cardiovascular procedures is associated with poorer outcomes. While underutilized, cardiac rehabilitation (CR) is guideline-recommended for patients undergoing cardiovascular procedures and may help mitigate the effects of frailty. This study evaluated the association between preprocedural frailty and CR use, as well as the interaction of frailty and CR use on 1-year mortality.

Medicare fee-for-service claims were queried for patients undergoing percutaneous or surgical revascularization or aortic valve replacement between July 2016 and December 2018. Patients who experienced mortality during the index admission or within 30 days of discharge were excluded. Patients were stratified into quartiles (Q1-Q4) using the validated claims-based frailty index (CFI). CR use was defined as attending any CR session within 1 year of discharge. Unadjusted comparisons and multivariable analyses were used to evaluate the relationship between frailty and CR use (CFI-Q4 versus CFI-Q1). An inverse probability treatment weighting model was used to determine if there was an interaction between CR, frailty, and 1-year mortality.

Overall CR use among the 501 049 beneficiaries was 37.7%; the average age was 75.9 years (SD, 7.3), and 37.0% were female. Increasing frailty was associated with decreased CR use (CFI-Q1: 49.7%, CFI-Q2: 42.2%, CFI-Q3: 35.3%, and CFI-Q4: 23.7%; P<0.001; adjusted odds ratio_{CFI-Q4 versus CFI-Q1}, 0.63 [95% CI, 0.62-0.64]). Unadjusted 1-year mortality was higher with increasing frailty (CFI Q1: 2.5%, CFI-Q2: 5.1%, CFI-Q3: 9.0%, and CFI Q4: 16.9%; P<0.001). After adjustment, the reduction in mortality associated with CR use was greater among frailer patients relative to less frail patients (CFI-Q4: 9.2% and CFI-Q1: 1.7%; P<0.001). CR use was associated with a significantly reduced association between CFI and 1-year mortality (P<0.001).

Preprocedural frailty is associated with lower CR use despite greater absolute benefits on 1-year mortality. Increasing CR use of frail Medicare beneficiaries may reduce 1-year mortality after cardiac interventions.